Senior Machine Learning Engineer (Audio Watermark Generation & Detection)

Senior Machine Learning Engineer 

(Audio Watermark Generation & Detection)

Songtradr Finland (EU)

VACANCY: FULL-TIME

Company Profile

Songtradr is the world's largest full-stack B2B music platform helping brands, content creators, and digital platforms find their voice and connect with audiences through music. 

Whether with a classic song or a trending tune, a global music strategy or a sonic identity, Songtradr helps translate ideas into powerful, ROI-driven solutions that always hit the right note.

The company’s fully integrated suite of products and solutions simplifies the process of finding, licensing and managing music across all formats. The result reduces cost and complexity, increases consistency and compliance, and elevates the full potential of music catalogs for users and rights holders alike.

Learn more at songtradr.com
and follow along on
Twitter, Instagram, Facebook or LinkedIn.

Job Description

We are seeking a talented and innovative Machine Learning Engineer to design, develop, and deploy robust systems for both generating and detecting specific audio watermarks. This role will involve researching, implementing, and optimizing solutions that may include proprietary adaptations of open-source watermarking algorithms (e.g., based on concepts similar to Audiowmark) or novel deep neural network (DNN) based approaches for both watermark embedding and retrieval. You will be responsible for the end-to-end implementation, from data generation and preprocessing to model/algorithm development, training, rigorous evaluation, and packaging the final solutions, with a strong focus on achieving high imperceptibility for generation and exceptional robustness against real-world audio distortions for detection.

As we continue to innovate in how music is transacted and protected, your expertise in developing cutting-edge audio watermarking systems will be pivotal in safeguarding assets and enabling new functionalities within our ecosystem.

Main Job Duties

• Develop & Implement Audio Watermarking Systems: Design, build, and refine algorithms for both generating and detecting robust and imperceptible audio watermarks using traditional and deep learning methods. This includes creating data generation pipelines with diverse audio augmentations.
• Audio Processing & Feature Extraction: Develop and apply advanced audio preprocessing techniques, feature extraction methods (e.g., Mel-spectrograms, MFCCs), and DSP algorithms.
• Model Training & Evaluation: Design, train, and rigorously evaluate watermark generation and detection models (including DNNs), focusing on performance metrics like imperceptibility, robustness, accuracy, and payload capacity.
• Research & Innovation: Stay updated on and contribute to advancements in audio processing, machine learning, and audio watermarking.
• Collaboration & Deployment: Work with cross-functional teams to integrate watermarking solutions into production, optimize algorithms, and document work.

Required Skills & Experience:

• Education: Master’s degree or equivalent practical experience in Computer Science, Electrical Engineering, Data Science, or a related field with a strong focus on Machine Learning and Signal Processing.
• Experience:
  • Proven experience in designing and developing machine learning models and/or signal processing algorithms for audio tasks.
  • Strong proficiency in Python and common machine learning frameworks (e.g., PyTorch, TensorFlow/Keras).
  • Solid programming skills in C++ for audio processing, algorithm implementation, and performance optimization.
• Knowledge:
  • Solid understanding of supervised learning, classification metrics, and model validation techniques.
  • Familiarity with various audio file formats and the nature of common audio distortions.
  • Good understanding of software development practices, including version control (Git).
  • Strong analytical and problem-solving skills.
  • Excellent communication skills in English.

Preferred Skills & Experience:

• Prior experience specifically in audio watermarking, audio fingerprinting, or audio forensics.
• Experience with advanced audio model architectures (e.g., 1D CNNs for waveforms, audio Transformers, CRNNs).
• Familiarity with the specific audiowmark library or similar audio watermarking tools.
• Experience with MLOps practices (model deployment, monitoring, containerization like Docker).
• Experience handling large datasets and working with cloud computing platforms (AWS, GCP).

What We Offer:

• A chance to work on challenging problems with cutting-edge AI/ML technology.
• Opportunity to build groundbreaking audio watermarking technology from the ground up with a choice of cutting-edge approaches.
• Direct impact on core company strategy related to content security and innovation.
• Work in a highly skilled, music-oriented environment with a passion for audio and technology.

How You’ll Fit Into Songtradr’s Culture 

CONSCIOUSLY COLLABORATIVE: You understand ‘team’ and collaborate with kindness and respect

HUMBLY CONFIDENT: You’re confident but are never arrogant

DILIGENTLY DRIVEN: You’re self-motivated, hard-working, and get results

INNOVATIVE SOLUTIONS: You’re positive, open, and passionate about finding innovative solutions

GOOD BUSINESS FOR ALL: You do what’s good for the business

Employment

Full time. What do you get in return? Inspiration, knowledge, career development, on top of our financial package. You’ll also be working with an international bunch of remarkable musically-infused individuals. 

On this note, please know that MassiveMusic is an equal opportunities employer. Applicants will not be excluded on the grounds of sex, gender reassignment, pregnancy, maternity, race, marital status, diversity of thought, disability, age, religion, belief, or sexual orientation. And if you need any specific adjustments to be made throughout our recruitment process, please feel free to let us know. 

If after reading this you know this is the perfect role for you, please apply via THIS LINK and make sure you include your resume and a brief summary of the professional achievement you are most proud of to date. 



Look forward to hearing from you!